Abstract
Remarkable progress has been made in the past few decades in various aspects of radiation therapy (RT). However, some of these promising technologies, such as image-guided online replanning and arc therapy, rely heavily on the availability of fast dose calculation. In this article, based on a popular dose calculation algorithm, the Collapsed-Cone Convolution/Superposition (CCCS) algorithm, we present a multi-FPGA accelerator to speed up radiation dose calculation. Our performance-driven design strategy yields a fully pipelined architecture, which includes a resource-economic raytracing engine and high-performance energy deposition pipeline. An evaluation based on a set of clinical treatment planning cases confirms that our FPGA design almost fully utilizes the available external memory bandwidth and achieves close to the best possible performance for the CCCS algorithm while using less resource. Compared with an existing FPGA design which aimed to accelerate the identical algorithm, the proposed design achieved 1.9X speedup by providing better memory bandwidth utilization (81.7% v.s. 43% of the available external memory bandwidth), higher working frequency (90MHz v.s. 70MHz) and less logic resource usage (25K v.s. 55K logic cells). Furthermore, it obtains a speedup of 20X over a commercial multithreaded software on a quad-core system and 15X performance improvement over closely related results. In terms of accuracy, the measured less than 1% statistical fluctuation indicates that our solution is practical in real medical scenarios.
- Ahnesjö, A. 1989. Collapsed cone convolution of radiant energy for photon dose calculation in heterogeneous media. Med. Phys. 16, 577--92.Google Scholar
Cross Ref
- Ahnesjö, A. 1997, Cone discretization for the collapsed cone algorithm. In Proceedings of the International Conference on the Use of Computers in Radiation Therapy. 114--116.Google Scholar
- Breitman, K., Rathee, S., et al. 2007. Experimental validation of the Eclipse AAA algorithm. J. Appl. Clinical Med. Phys. 8, 2, 76--92.Google Scholar
Cross Ref
- Chauvie, S., Dominonic, M., Marinic, P., Stasia, M., Piab, M. G., and Scielzoa, G. 2003. Monte Carlo dose calculation algorithm on a distributed system. Nuclear Physics 125, 159--163.Google Scholar
Cross Ref
- Bowen, M. 1998. Handel-C Language Reference Manual. Celoxica Ltd, Abingdon, UK.Google Scholar
- Chen, Q., Chen, M., and Lu, W. 2011. Ultrafast convolution/superposition using tabulated and exponential kernels on GPU. Med. Phys. 38, 3, 1150--1161.Google Scholar
Cross Ref
- Detrey, J. and Dinechin, F. D. 2003. A VHDL library of LNS operators. In Proceedings of the 37th Asilomar Conference on Signals, Systems and Computers. 2227--2231.Google Scholar
- Fraass, B. A., Smathers, J., and Deye, J. 2003. Summary and recommendations of a National Cancer Institute workshop on issues limiting the clinical use of Monte Carlo dose calculation algorithms for megavoltage external beam radiation therapy. Med. Phys. 30, 3206--3216.Google Scholar
Cross Ref
- de Greef, M., Crezee, J., van Eijk, J. C., Pool, R., and Bel, A. 2009. Accelerated ray tracing for radiotherapy dose calculations on a GPU. Med. Phys. 36, 9, 4095--4192.Google Scholar
Cross Ref
- Guo, Z., Buyukkurt, B., Najjar, W., and Vissers, K. 2005. Optimized generation of data-path from C codes for FPGAs. In Proceedings of the Conference on Design, Automation and Test in Europe (DATE'05). 112--117. Google Scholar
Digital Library
- Gu, X., Choi, D., Men, C., Pan, H., Majumdar, A., and Jiang, S. B. 2010. GPU-based ultra-fast dose calculation using a finite size pencil beam model. Phys. Med. Biol., 54, 20, 6287--6297.Google Scholar
Cross Ref
- Hissoiny, S., Ozell, B. and Després, P. 2009. Fast convolution-superposition dose calculation on graphics hardware. Med. Phys. 36, 1998--2005.Google Scholar
Cross Ref
- Jacques, R., Taylor, R., Wong, J., and Mcnutt, T., 2008. Towards real-time radiation therapy: GPU accelerated superposition/convolution. In Proceedings of the High-Performance Medical Image Computing and Computer Aided Intervention Workshop.Google Scholar
- Lu, W., Olivera, G. H., Chen, M. L., Reckwerdt, P. J., and Mackie, T. R. 2005. Accurate convolution/super-position for multi-resolution dose calculations using cumulative tabulated kernels. Phys. Med. Biol 50, 655--680.Google Scholar
Cross Ref
- Luu, J., Redmond, K., Lo, W., Chow, P., Lilge, L., and Rose, J. 2009. FPGA-based Monte Carlo computation of light absorption for photodynamic cancer therapy. In Proceedings of the IEEE Symposium on Field-Programmable Custom Computing Machines (FCCM'09). 157--164. Google Scholar
Digital Library
- Mackie, T. R., Bielajew, A. F., Rogers, D. W. O., and Battista, J. J. 1988. Generation of photon energy deposition kernels using the EGS Monte Carlo code. Phys. Med. Biol. 33, 1--20.Google Scholar
Cross Ref
- Maspradakis, M., Morrison, R. H., Richmond, N., and Steele, A. 2003. Experimental verification of convolution/superposition photon dose calculations for radio therapy treatment planning. Phys. Med. Biol. 48, 2873--2893.Google Scholar
Cross Ref
- McGary, J. E. and Boyer, A. L. 1997. An interactive, parallel, three-dimensional fast Fourier transform convolution dose calculation using a supercomputer. Med. Phys. 24, 519--522.Google Scholar
Cross Ref
- McNutt, T. R. 1999. Dose calculations collapsed cone convolution and delta pixel beam, White Paper Publication, ADAC Laboratories (Philips), Milpitas, CA.Google Scholar
- Reckwerdt, P. J. and Mackie, T. R. 1992. Superposition/convolution speed improvements using runlength. Med. Phys. 19, 784.Google Scholar
- Wang, C., Luan, S., Tang, G., Chen, D. Z., Earl, M. A., and Yu, C. X. 2008. Arc-modulated radiation therapy (AMRT): A single-arc form of intensity-modulated arc therapy. Phys. Med. Biol. 53, 22, 6291--6303.Google Scholar
Cross Ref
- Whitton, K., Hu, X. S., Yu, C. X., and Chen, D. Z. 2006. An FPGA solution for radiation dose calculation. In Proceedings of the 14th Annual IEEE Symposium on Field-Programmable Custom Computing Machines, 227--236. Google Scholar
Digital Library
- Zhou, B., Hu, X. S., Chen, D. Z., and Yu, C. X. 2007. Hardware acceleration for 3-D radiation dose calculation. In Proceedings of the 18th IEEE International Conference on Application-specific Systems, Architectures and Processors. 290--296.Google Scholar
- Zhou, B., Hu, X. S., Chen, D. Z., and Yu, C. X. 2009. A multi-FPGA accelerator for dose calculation in radiation therapy. In Proceedings of the 51st Annual Meeting of the American Association of Physicists in Medicine.Google Scholar
- Zhou, B., Hu, X. S., Chen, D. Z., and Yu, C. X. 2009. A multi-FPGA accelerator for radiation dose calculation in cancer treatment. In Proceedings of the 7th IEEE Symposium on Application Specific Processors. 70--79.Google Scholar
- Zhou, B., Yu, X. S., Chen, D. Z., and Hu, C. X. 2010. GPU-accelerated Monte Carlo convolution/superposition implementation for dose calculation. Med. Phys. 37, 11, 5593--5603.Google Scholar
Cross Ref
- Zhou, B., Hu, X. S., Chen, D. Z., and Yu, C. X. 2010. Dose calculation accelerating: A comparison study of GPU and FPGA based on collapsed cone algorithm. In Proceedings of the 52nd Annual Meeting of the American Association of Physicists in Medicine.Google Scholar
Index Terms
Accelerating radiation dose calculation: A multi-FPGA solution
Recommendations
A multi-FPGA based platform for emulating a 100m-transistor-scale processor with high-speed peripherals (abstract only)
FPGA '10: Proceedings of the 18th annual ACM/SIGDA international symposium on Field programmable gate arraysThis paper describes a multi-FPGA based platform for emulating the Loongson-2G micro-processor on different mother boards. This platform is developed targeting at verification and evaluation of the Loongson-2G micro-processor, which is the next ...
An FPGA Solution for Radiation Dose Calculation
FCCM '06: Proceedings of the 14th Annual IEEE Symposium on Field-Programmable Custom Computing MachinesRadiation dose calculation is an important step in the treatment of cancer patients requiring radiation therapy. It ensures that the physician prescribed dose agrees with the dose delivered to the patient. Current methods use software implementing ...
Accelerating 3D CNN-based Lung Nodule Segmentation on a Multi-FPGA System
FPGA '19: Proceedings of the 2019 ACM/SIGDA International Symposium on Field-Programmable Gate ArraysLung nodule segmentation is one of the most significant steps in many Computer Aided Detection (CAD) systems used for lung nodule identification and classification. Three-dimensional convolutional neural networks (3D CNNs) have become a promising method ...






Comments